---
title: "skills"
description: "Skills for interface design, frontend implementation, repository review, writing, and delivery."
canonical: https://agentpluginsdirectory.com/plugins/skills
last-updated: 2026-10-21
---

# skills
Skills for interface design, frontend implementation, repository review, writing, and delivery.
- Slug: skills
- Publisher: Claudia Nathan
- Repository: https://github.com/claudialnathan/skills
- Manifest: plugin.json
- Version: 1.0.1
- License: MIT
- Category (editorial): productivity
- Skills: 15 (designer, explain-system-flows, improve-composition, improve-layout, improve-motion, inspect-web, openreview, optimistic-ui, quality-audit, saltintesta, ship, use-browser, video-to-ascii, web-launch-checklist, zoom-out)
- MCP servers: 0
- Stars: 1
- Repository created: 2026-07-14
- Repository last pushed: 2026-10-07
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/skills
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What skills does, in the publisher's words

Agent skills that add meaningful value beyond what an unaided model does out of the box, for work that matters to me.

A frontier model is very competent at most things, but its default in any domain is the competent average version of the thing. Each skill here is an effort to drag the model off that median in a chosen direction: narrow, stack-specific disciplines with explicit boundaries and proof obligations.

They carry my taste and intent, stack-specific decisions, failures observed in real work, and the evidence I require before calling work done. The agent gets more specific instructions for the work I care about without putting all of that back into every prompt.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/claudialnathan/skills/HEAD/README.md

## Skills

- designer: Apply whenever building, finishing, or reviewing UI in a React + Tailwind + shadcn project. Every taste call carries its one-line reason, scale, easing, duration, radius, shadow, quantitative where possible, named in precise vocabulary; the finish details get proposed unprompted where they fit, f…
- explain-system-flows
- improve-composition: Audits and improves a web interface as one composition across product intent and states, foundations, primitives, canonical UI, reusable product compositions, routes, and component catalogs. Use when asked to audit, reconcile, standardize, or make coherent a design system, components page, UI archi…
- improve-layout: This skill should be used when the user asks to audit, fix, build, or improve page layout, app shells, Kanban boards, responsive and touch behavior, swipeable or paging columns, grids, sidebars, full-height screens, breakout content, fluid type or spacing, element alignment, or narrow-width overflo…
- improve-motion: This skill should be used when the user asks to audit, review, fix, simplify, refactor, remove, add, or improve UI animation, motion, transitions, page or view transitions, micro-interactions, gestures, scroll effects, layout animation, enter/exit behavior, easing, springs, or animation performance…
- inspect-web: This skill should be used when the question is how a live web page achieves something, or why it looks or behaves wrong, an animation, view transition, layout, type scale, colour scheme, or load speed, on our own page or a third-party reference, whether to reproduce the effect or to diagnose it.…
- openreview: This skill should be used when the user asks to audit, review, improve, fix, plan, reconcile, or implement Next.js framework work with OpenReview. It runs a separately provisioned OpenReview scanner across a repository or current-HEAD diff, vets deterministic diagnostics and high-leverage framework…
- optimistic-ui: Default to optimistic-UI patterns where perceived speed matters. Render from cached state immediately, mutate optimistically with rollback, don't gate render on session validation, and never let heavy synchronous work sit between an input and the next paint. Applies to mutation handlers (save / edi…
- quality-audit: This skill should be used for a whole-repository, stack-aware quality audit of a JavaScript or TypeScript web project; pre-ship risk review; launch-readiness checkup; or scheduled read-only scan. It detects applicable framework, UI, server, data, security, accessibility, performance, state-integrit…
- saltintesta: Apply whenever writing anything a person will read: prose in any form, from a README to an article to a message. Articulates ideas in as few good words as possible; grills the user when the idea is missing. Use when asked to tighten, shorten, or make it sound less AI.
- ship
- use-browser
- video-to-ascii
- web-launch-checklist
- zoom-out

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.